w3c
Integrate DID Traits Schema
As discussed in the W3C CCG call on Jan 7, the proposal was made to the DID Traits schema definition be integrated with the DID Extensions registry.
- The DID Traits specification defines traits of DID methods.
- Traits are described in JSON. There are a number of example assessments for well-known DID methods in the spec repository: https://github.com/decentralized-identity/did-traits/blob/main/methods/
- A JSON Schema exists that is referenced by the traits definitions https://identity.foundation/did-traits/schemas/v0.8.0/traits.json
Current status of the specification:
- The specification has been approved by the working group.
- The specification has NOT YET received final approval by DIF.
Before creating a PR, I would like to align with you on how the integration shall done, e.g. an additional file in the methods folder methods/<DID>-traits.json or methods/traits/<DID>.json. Furthermore, it might be worth thinking about ways to visualize traits. See https://identity.foundation/did-traits/v0.8.0/#comparison-of-did-methods for an example.
@msporny
This was discussed during the did meeting on 03 April 2025.
View the transcript
w3c/did-extensions#619
manu: Happy to have a discussion about DID Traits and Rubric... I do think we could integrate the traits into the DID Extensions quite easily by adding a 'traits' array with string identifiers for each trait and making it optional for DID Method registrants to fill that out by themselves. We can provide a table of trait identifiers to explanations about what the trait means.
ottomorac: Next week is IIW, so we won't have a call until April 17th.
<ottomorac> m2gbot, link issues with transcript
<m2gbot> Something wrong happened: Failed loading minutes from https://
This was discussed during the #did meeting on 17 April 2025.
View the transcript
DID Rubric & Traits discussion
<ottomorac> w3c/
<markus_sabadello> See discussion here about JSON and JSON-LD for DID Resolution Result: w3c/
ottomorac: I'd like to introduce Jan Christoph Ebersbach
JCE: thanks, we continued to work on the DID Traits effort at DIF
… the idea was to provide a list of relevant features or trait that DIDs use, and record them in a machine-readable format
<ottomorac> https://
JCE: we have different traits like Self-Certifying Identifiers, etc
… we thought about how are we going to use this list of traits. and one way we found useful is to create an overview
… to compare different DID Methods
… so the did-traits link has a comparison table that gives an idea of how they can be used
… and I raised an issue in the DID Extensions repo to discuss possible integration of Traits into the extension registry
… so that method authors can record traits in their registration entry
manu: thank you JCE and ottomorac for this wonderful work
… one way we can do the integration is - when DIDs are registered in the Extensions list, we could add a field to the registration, that's quite literally reuses the JSON Schema you provided
… so that when people go in and register, they can also, through a property called 'traits', list all the traits that apply
… which would allow them to set booleans to true, or whatever it is
… that might be the easiest integration path
… there are other secondary things like - do we want people to be able to select traits, and filter the extensions list?
… but obv we need data to do that
… the question I have for you JCE and Otto is -- how ready to go do you feel Traits are?
… I'm sure we'll add them over time, etc. do we point elsewhere for the Traits schema? does the WG have to review all the traits and agree?
… those are the open question I have. but the integration feels straightforward and mechanical to me
JCE: the main thing is - we received approval from the DIF WG for the version 0.8
… we're currently pursuing v1.0
… we did receive some feedback, and there will be additional changes to the list
… in general, the Traits are represented as booleans, we tried to stick with Traits that are very technical (and therefore CAN be represented with booleans, and still convey value and meaning)
… I'd say we've covered a lot of ground with Traits already. and of course want additional feedback
… in the future, the list will be extended, like you said Manu. a good way would be to go through DIF for extensions
… I'm not bound to keeping the Traits spec at DIF
… maybe it'd be easier to integrate with W3C, to have under one roof
<Zakim> JoeAndrieu, you wanted to suggest we should build on the rubric
JCE: but that's more a question for you all
JoeAndrieu: I'm a bit confused by the tone of conversation -- we passed a resolution to explore how to integrate Traits with the Rubric
… we had a good conversation. i think the DID Traits do fit
… for each trait that exists, we can create a Criteria in the Rubric
… I would be opposed to putting DID Traits in to the Extensions registry, and ignoring the work we've done on the Rubric
… I'd like to see how we can integrate the two, before adding to the Extension
Wip: with chair hat off, I agree with Joe
… for example, one of the trait is Multi-Signature Verification method. and if I understand, ANY DIDs can support multi signature
… at least that's how I understand it
… and that highlights a challenge with the DID Traits approach
… that there's just boolean flags to check,
manu: couple of thoughts
… seems like there's a desire to at least move the Traits work into this WG in some way, and integrate it in some way
… whether through the Rubric, or as items in the DID Extensions registry
… so we just need to figure out how it happens
… I agree with Will's point -- all these different verification methods, every DID method should support all of thsoe,
<ottomorac> I will note that I agree that this needs to evolve into W3C as well, not sure what form though
manu: so those boxes will always be checked
… I don't view this as throwing out whatever work was done with the Rubric
… I'm most interested in a way for people to filter DID Methods based on features / criteria, at least on a high level
… there's some criteria we just can't make into a checkbox. but some, we can!
… so, any kind of downselecting / filtering mechanism would be better than just a giant list of 200+ did methods
… so, question is - what's the next step for the group?
… does DIF have any interest in handing some of this over to W3C? IP-wise etc
… I don't think there'd be much IP concerns -- it was incubated at DIF, can be handed to the WG, etc
… again, thing I care most about is the end result -- will people have a better way of navigating the DID methods
… otherwise, it'll be overwhelming / not very useful
markus_sabadello: I think both DID Traits and the Rubric are very useful tools for this kind of filtering
… therefore I'd be very excited to integrate it with the Extensions registry
… I admit I don't remember exactly the resolution we passed previously about integrating Traits with Rubric
… I think one of the aspects was - traits has boolean flags, whereas Rubric was more extensive
… I wonder if this integration is -- I understand what Joe is saying about the order of procedure
… but I wonder if there's any downside to integrating Traits with Extensions in parallel to integrating Rubrics with traits and extensions
… regarding contributions from DIF - maybe JCE can clarify
<Wip> These are our resolutions - https://
ottomorac: just to acknowledge the point, Will just pasted the specific link to the resolution
… this was discussed, I remember,
JCE: so, two things - with regards to data collection, the right place is the DID Extensions Registry
… and where the spec is maintained -- it's not an issue to move it away from DIF, if it's better to move it to W3C, no prob, we'll talk internally at DIF and discuss it, but feels like no obstacles
… one additional point regarding different traits -- we do have a definition of each trait. so if you're not 100% sure what a trait means, there's a more detailed description. if you have questions, please raise issues on the tracker
… next step wise, to me it feels like we should wait for the 1.0 approval of DID Traits by DIF
… and then we can pursue the integration officially with Extensions registry etc. and in the meantime, we can prepare a PR so that we can start experimenting with the integration
… maybe also work on some display / filtering code
<JoeAndrieu> I would oppose integration that didn't enable Rubric evaluations in the same manner
ottomorac: thanks all, we'll continue discussion in the next call
